OHDSI Home | Forums | Wiki | Github

Management of investigational drugs / agents

Vocabulary Users
#1

Hello Vocab Folk:

At N3C we are wondering how the OHDSI community manages codes / coding for investigational drugs?
One of the concerns is the “naming” life cycle of an investigational agent in a longitudinal dBs it progresses through approval gateways. Additionally, in blinded studies, what (if any) provision is there within the OMOP dB to indicate the possibility (not completely confirmed) that a patient may have been exposed to an investigational agent, the possibility existing said patient could have been in a placebo arm?

I searched the forum and was not finding comments: if this has been previously discussed - if you could kindly point me in the right direction, I would appreciate that

Many thanks in advance for your assistance

Davera Gabriel
davera@jhu.edu

#2

There is no magic solution.

If you db engine is fast, you can rebuilt target OMOP data from scratch daily or weekly. So if a concept (drug) is later classified better, your ETL will simply use latest local or Athena mappings and result in more up to date target OMOP data.

The Study WG did discuss this problem.

I have a question to you in turn: Can you describe if N3C will put into same database study data with routine EHR data (even if it is not part of core study design)? How many COVID studies N3C expects to work with (do you have any (to be mapped to OMOP) today?)

#3

Thank-you for your speedy response, Vojtech! I think that is possible, albeit not currently happening right now. CTSAs have EHRs and also conduct RCTs in their service populations - so co-mingling these data is not inconceivable. Additionally, I am aware that exposure to investigational drugs can be components of EHR records, and as such those identifiers require management in EHR implementations, especially regarding the “name lifecycle” in longitudinal records.

Dr Chute is planning to discuss the issue of investigational drugs in our Data Ingestion & Harmonization public meeting this Wednesday (11am Eastern) So it would be awesome if you might attend and engage in the discussion he would like host.

t